Kimi uncensored API Quickstart Guide

Authentication and Base URL

The endpoint for our uncensored API is HTTPS://api.kimiuncensored.com/v1. Pass your key in the Authorization header using the Bearer scheme. This setup works with any client that supports OpenAI-compatible authentication.

• Sign up via Google or email to receive your key immediately.
• No phone number or payment method is required for the initial trial credit.
• Your API key is unique to your account and replaces any previous key if you regenerate it.

Copycurl https://api.kimiuncensored.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "List three ways to optimize PostgreSQL query performance."}]
  }'

POST /v1/chat/completions

Send a JSON payload to the chat completions endpoint to generate text. The model ID is simply "uncensored". You can control the output style using parameters like temperature and top_p.

• Set max_tokens to limit the response length.
• Use the stop parameter to halt generation at specific sequences.
• The system supports standard OpenAI request structures.

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.kimiuncensored.com/v1", api_key="YOUR_KEY")

resp = client.chat.completions.create(
    model="uncensored",
    messages=[{"role": "user", "content": "Summarise this thread without softening it."}],
)
print(resp.choices[0].message.content)

Streaming Responses (SSE)

Enable streaming by setting stream: true in your request. The API returns Server-Sent Events (SSE) containing partial text chunks. This reduces perceived latency for interactive applications.

• Token usage data is included in the final chunk of the stream.
• Handle network interruptions gracefully as streams may break.
• This method is ideal for real-time user interfaces.

Copystream = client.chat.completions.create(
    model="uncensored",
    messages=[{"role": "user", "content": "Tell the story in second person."}],
    stream=True,
)
for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

JSON Mode and Function Calling

Force structured output by setting response_format to {"type": "json_object"}. This ensures the model returns valid JSON, which is crucial for parsing data in automated workflows.

• Define tool definitions in the tools array for function calling.
• Use tool_choice to control when the model calls functions.
• Validate the JSON response locally before processing.

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.kimiuncensored.com/v1", apiKey: process.env.API_KEY });

const resp = await client.chat.completions.create({
  model: "uncensored",
  messages: [{ role: "user", content: "Draft a villain monologue for my game." }],
});
console.log(resp.choices[0].message.content);

Rate Limits and Constraints

Each API key is limited to 300 requests per minute and 8 concurrent requests. The maximum request body size is 8 MB. Exceeding these limits results in a 429 status code.

• A 401 error indicates an invalid or expired API key.
• A 402 error means you have insufficient prepaid credit.
• Errors and refusals do not consume your token balance.

Context Window and Model Details

The model supports a context window of 100,000 tokens for both prompt and completion combined. If you do not specify max_tokens, the default output limit is 2,048 tokens. This allows for long-context processing while maintaining response speed.

• Input tokens are charged at $0.15 per 1M tokens.
• Output tokens are charged at $0.50 per 1M tokens.
• Prepaid credit never expires and is topped up via crypto.